Synthesis of Sulfur-Grafted Chitosan Biopolymers and Improvement to Their Sorption of Silver Ions from Different Aqueous Solutions

Abstract

A functionalized chitosan thiourea composite (CH-TU) was successfully synthesized using formaldehyde as a crosslinking agent for enhancing silver recovery from different aqueous solutions. Comparison sorption studies with a non-functionalized composite (CH-F) as a reference material were conducted. Grafting led to an improvement in the sorption performances, i.e., 0.763 mmol Ag g⁻¹ for CH-F vs. 2.125 mmol Ag g⁻¹ for CH-TU. The pseudo-first-order rate equation (PFORE) was fitted to the sorption kinetics at saturation times of 40 and 30 min for CH-F and CH-TU, respectively, while the sorption isotherms were fitted with Langmuir and Sips equations for both sorbents. Fourier transform infrared (FTIR), scanning electron microscopy (SEM), energy-dispersive X-ray analysis (EDX), nitrogen sorption–desorption isotherms (BET-surface area), elemental analysis, thermogravimetric analysis (TGA) and pH of the zero charge (pH_pzc) were used to characterize and investigate the sorption mechanism. Sorption was performed three times to check the reproducibility, while the sorption performances were stable over 20 cycles, with a limited decrease in performance (5 and 3% for CH-F and CH-TU, respectively). Nitric acid solution (0.3 M) was efficient for desorbing the adsorbed metal ions. The grafted sorbent with thiourea is considered as a promising tool for recovering Ag(I) from acidic waste leachate derived from waste spent films.

1. Introduction

Silver is considered as a precious metal and is used in several industries, especially in the jewelry industry. Some of its alloys, i.e., silver–copper, are used in computer manufacturing. Silver is used in high-tech industries (i.e., electrical connectors), electronics, products for health care, photovoltaics cell production [1,2], button cells [3], and several nanomaterial industries [4,5,6,7]. Silver is used in light reflection industries for mirrors and photographic films (and X-ray films) because of its sensitivity to light and in photochromic lens manufacturing [8]. Most silver is derived as a sub-product from the refining of cadmium, zinc and copper [9]. Some hydrometallurgical procedures consume a huge amount of energy, which consequently increases the cost of recovery or decreases the efficiency of recovery of the desired metals; this results in toxic by-products which are a secondary pollution source.

Silver is considered as a toxic material for aquatic life [10], including fish. Many countries list silver as a pollutant for water bodies [11]. Several technologies have been used in the procedure of silver extraction from aquatic media, wastewater and some other environments. Among these technologies are biosorption, biohydrometallurgy, reverse osmosis, leaching, cementation, ion exchange, adsorption, electrochemical deposition and ultrafiltration [5,12,13,14,15,16]. This is one of several reasons that encouraged researchers to develop processes to recover silver from diluted media, decreasing its effect on humans and the environment. It is noteworthy that there is a serious need to find a low-cost, effective and sustainable method for recovering silver (with a high efficiency) from leachates, wastewater, effluents, waste materials or other secondary resources. Several techniques have been used for reducing the silver concentration; i.e., electrodeposition methods have been used [1,17,18] and solvent extraction techniques have been used for recovering silver from highly concentrated solutions [19,20,21]. Extraction from low-grade solutions is performed through chelating composites [16,22], ion-exchange resins [23,24,25,26] and biosorbents [27,28,29]. Green technologies derived via microbial metabolism have effectively been used in the treatment of heavy metals and wastewater [30,31,32]. Sulfur-based sorbents (thions, thiols or heterocyclic compounds) show a high affinity toward silver ions [22,23,33,34,35]; due to the sulfur reagent compounds, which shows a high selectivity for silver [36,37,38,39,40,41], this is based on hard and soft acids and bases (HSAB) theory [42]. Selectivity has been shown by sulfur-containing polymers, i.e., dithiocarbamate [43], mercaptobenzimidazole [35] and thiol derivatives [44,45].

The easy functionalization of biopolymers by grafting new groups for either enhancing the sorption capacities or improving the selectivity and other physicochemical properties (for specific metals or groups of metals) has received the attention of the scientific community. This can be achieved physically using beads, fibers or hollo fibers to improve the sorption kinetics and physical properties. Coupling of different biopolymers has been achieved physically or chemically by crosslinking [46,47,48,49,50,51]. Several functional groups have been grafted for this purpose, such as amidoxime groups [52], dithiooxamide [22], quaternary ammonium derivatives [53], sulfonic moieties [54], amino acid derivatives [55,56], amine and thion [57], amino sulfonic acid [58], phosphorus groups [59,60] heterocyclic groups [61,62,63], and the green synthesized materials [64].

Chitosan (amino-polysaccharide) is well known as one of the most abundant biopolymers (produced commercially from shrimp shells). It has been used as a base material for the efficient removal of environmental pollution, including dyes and radioactive and heavy elements. In addition, it is used in the medical field in drug delivery. The presence of amines in the structure allows it to be dissolved in acidic media (through protonation), and it can be physically modified easily (through beads, fibers or hollo fibers [65]) or chemically [66,67,68] (though grafting new functional groups or compounds) to enhance the sorption and/or improve the kinetics. The most interesting groups that have been grafted on chitosan particles are polyamine moieties [69,70], carboxylic/carboxylate derivatives [71,72] and phosphonic [73,74] and aminophosphonic [75,76] groups.

This work is concerned with developing a sustainable extraction process with a low cost and high efficiency for recovering silver using a functionalized chitosan composite biopolymer through a reaction with thiourea and formaldehyde as a coupling agent in water media (without using a toxic solvent). This work was performed due to the disadvantages of most recent methods in the field of silver recovery (including a high consumption of energy, a low efficiency and the production of toxic by-products). The sorbents were first synthesized and characterized through a broad variety of analysis tools, i.e., FTIR, TGA-DTA, SEM-EDX, elemental analysis, titration and BET (sorption–desorption isotherms of nitrogen). The second part of the study details sorption optimization through different kind of tests, such as pH effect, sorption isotherm, sorption kinetics and selectivity tests (from multicomponent solution); this was conducted for the optimization of Ag(I) ion sorption from synthetic solutions before application in a complex real solution. The last part of the work uses the optimum conditions from the previous step in an experiment on real effluents from secondary waste resource materials.

2. Materials and Methods

2.1. Materials

Chitosan powder (25% acetylation degree), thiourea (>99%), formaldehyde solution (37 wt.% in H₂O), calcium sulfate dihydrate (CaSO₄.2H₂O, >97%), silver nitrate (AgNO₃; >99%), magnesium sulfate heptahydrate, (MgSO₄·7H₂O, ≥99.5%), zinc sulfate heptahydrate (ZnSO₄·7H₂O, 99%) and cadmium sulfate (CdSO₄, ≥99.0%) were purchased from Sigma Aldrich, Merck KGa and Darmstadt, Germany. Aluminum sulfate octahydrate (Al₂(SO₄)₃·8H₂O, >99%) and sodium sulfate decahydrate ((Na₂SO₄·10H₂O, ≥99.0%) were provided by Guanghua Sci-Tech Co.-Ltd, Guangdong, Shantou, China.

2.2. Synthesis of Adsorbents

2.2.1. Synthesis of Chitosan Formaldehyde Particles (Reference Material)

A total of 4 g of chitosan particles was dissolved in 150 mL of a 7% (w/w) acetic acid solution. The mixture was transferred into a 250 mL three-necked round-bottom flask. A total of 5 mL of formaldehyde solution was added drop-wise to the mixture with continuous stirring, before the flask was equipped with a condenser and a thermometer to adjust the temperature to 90 °C, and the reaction was continued for a further 5 h to produce the crosslinked chitosan composite. The precipitate product was filtered and washed with water and acetone (to remove the unreacted materials), then dried at 60 °C overnight to produce non-functionalized chitosan particles (CH-F), as shown in Scheme 1.

2.2.2. Synthesis of Functionalized Chitosan Thiourea Particles

Four grams of chitosan particles was dissolved in 150 mL of 7% acetic acid solution (as in the previous step). A total of 4 g of thiourea was added to the mixture with continuous stirring until it dissolved, followed by 10 mL of formaldehyde solution, added drop-wise (in 10 min). The flask was equipped with a condenser and a thermometer, and the reaction was continuously stirred under reflux (at 90 °C) for 5 h. The produced precipitate was filtered and washed with water and acetone before being dried at 60 °C for 12 h to produce functionalized chitosan thiourea particles (CH-TU), as shown in Scheme 2.

The expected synthesis mechanism was assigned as an elimination reaction of oxygen from formaldehyde with two protons from the amine groups (of either chitosan and/or thiourea, depending on the sorbent), as shown in Scheme 3.

2.3. Characterization

Fourier transform infrared (FT-IR) spectra of the dried samples, which were used for verifying the structure and the specification of functional groups (the grafted groups), were measured after introducing the samples into a KBr disc using an IRTracer-100 FT-IR spectrometer, Shimadzu, Tokyo (Japan); the resolution of the equipment was set to 4 cm⁻¹. Nitrogen sorption–desorption isotherms were used for detecting the porosity and surface area using a TriStar-II system, Norcross, GA, USA, at −77 K. The samples were first swept for 4 h under N₂ gas at 110 °C. Thermal decomposition of the composites was achieved through a Netzsch STA 449 F3-Jupiter, NETZSCH, Gerätebau GmbH, Selb, Germany. Experiments were performed under nitrogen, with 10 °C min⁻¹ as the temperature ramp. Morphological studies were performed using scanning electron microscopy (SEM) of the sorbents, achieved using a Phenom ProX-SEM, Thermo-Fisher Scientific, Eindhoven, The Netherlands, with an accelerating voltage of 15 Kv. Energy-dispersive X-ray analysis (EDX tool) was used for verifying the semi-quantitative analysis. The total charge of the surface at different pH values was studied using pH of the zero charge (pH_pzc) through the pH-drift method [77]. The pH was adjusted using an ionometer (S220 Seven; Mettler-Toledo, Shanghai (China)). The collected solutions from the experiments (before and after sorption) were filtered (by 1.2 µm filter membranes) before analysis using an inductively coupled plasma atomic emission spectrometer (ICP-AES, Agilent Technologies, Santa Clara, CA, USA). The elements in the solutions (residues after sorption and the original solutions) were measured using an ICPS-7510, Shimadzu, Tokyo (Japan).

2.4. Sorption and Application Experiments

The batch method was used for describing the sorption performances of functionalized and non-functionalized sorbents. It was performed under agitation with a velocity of 200–220 rpm as follows: a solution with an initial concentration (C₀, mmol Ag L⁻¹) at an initial pH value (pH₀) of a certain volume (L) was made up, containing a certain amount of sorbent (m, g) which is linked to the sorbent dose (Equation (1); g L⁻¹). The samples collected from kinetics (after a fixed time (C(t), mmol Ag L⁻¹)) or after equilibrium (C_eq, mmol Ag L⁻¹) from isotherm, pH, or selectivity experiments were analyzed via icp after filtration, as mentioned before. In the sorption isotherms, different initial metal concentrations were detected (0.09 and 4.8 mmol Ag L⁻¹) in the equilibrium residue. Tables S1 and S2 report the relevant kinetics and isotherm equations. The sorption selectivity was investigated using equimolar concentrations of different categories of metal ions (expected to be co-existing) with Ag(I) in the leachate solution. This experiment was performed at different pH values. The desorption kinetics (determined from the loaded samples in kinetics experiments) of adsorbed Ag ions in an acidic 0.3 M nitric acid solution were determined. Sorption recycling was performed in five cycles, while a rinsing step (using demineralized water) was performed between each run. The sorption capacity was calculated using a mass balance equation, Equation (2)

SD = m/V

(1)

Q = (C₀ − C_eq) × V/m

(2)

Experiments were performed in triplicate for testing the reproducibility and were plotted using error bars for the average of the three experiments. The mathematical equations for kinetics and isotherms are listed in the Supplementary File; see Tables S1 and S2 for the kinetics and isotherms equations, respectively.

The polymetallic-contaminated solution containing Ag(I) ions was derived from acidic leaching of waste material. Sorption was investigated using the batch method (200 rpm) at 21 °C with S.D (0.6 gL⁻¹) for 10 h. The initial and final metal concentrations were measured using ICP-AES to determine the sorption capacities, distribution ratios and selectivity.

3. Results

3.1. Characterization of Synthesized Sorbents

3.1.1. FTIR Analysis

Different types of functional groups were detected in the FTIR spectra (Sciencetech Inc., London, ON, Canada), verifying the structure and the successful grafting of new groups. The analysis was performed on the pristine biopolymer and the biopolymer after functionalization with thiourea moieties.

The synthesized chitosan composites, non-functionalized (CH-F) and after grafting of thiourea (CH-TU), were characterized by a variety of different groups; i.e., OH and NH were found in both composites and S=O and S-H were found in the functionalized particles, as shown in Figure 1. The peaks at 3429 and 3412 cm⁻¹ for CH-F and CH-TU, respectively, were assigned to ν_O-H and ν_N-H [78,79,80,81]. These peaks appeared with a low intensity and almost disappeared after adsorption of Ag(I) (this points to the sharing of NH and OH in the sorption mechanism), while they returned to their original state after desorption. The peaks around 2920 to 2940 cm⁻¹ were assigned to the ν_C-H of aliphatic bonds [82,83]. The peaks in the region of 2300–2400 for CH-F are related to the standard error of the equipment. The new peak that appeared after grafting of thiourea at 2495 cm⁻¹ is related to the ν_C=S of the grafted moieties [84], which verifies the successive grafting of thiourea. The peak at 2079 cm⁻¹ was assigned to ν_SH (through tautomerization) [85]; these two peaks disappeared after sorption, verifying the binding of metal ions [86], and were restored, with small shifts for the desorbed composite in the peaks at 2499 and 3135 cm⁻¹. The broad band at 1624 cm⁻¹ for CH-TU compared to weak band at 1631 cm⁻¹ for CH-F highlights the successive grafting of thiourea; these peaks are related to NH groups [87]. Some peaks shifted or appeared with a low resolution after metal sorption, leading us to conclude that they were involved in the binding mechanism [88]. Among these peaks are the OH and NH peaks of both sorbents and the C=S and SH peaks for CH-TU. Table S3 shows some of the interesting peaks of both sorbents after loading and after 20 cycles of sorption and desorption.

3.1.2. TGA Analysis

The thermal decomposition of both sorbents (CH-F and CH-TU) was investigated using TGA analysis, as shown in Figure 2, while the DrTGA data are reported in Figure S1. The two sorbents had similar trends in their decomposition profile shapes, while they showed different results in the overall loss percent and the final residues. They had three loss profiles, which are described in the following.

(a) The first step was assigned to the release of absorbed water, showing at 195.2 and 228.3 °C for CH-F and CH-TU, respectively; the loss in weight at this step is around 19.53 and 12.66% for both sorbents, respectively.

(b) The second step is concerned with the degradation of functional groups (amine and hydroxyls in the CH-F sorbent and amine, hydroxyls and thiol/thion in the CH-TU sorbent), as well as depolymerization of the sorbents. This step exhibited a weight loss of 55.9 and 50.12% for both sorbents, respectively.

(c) The last step is concerned with the degradation of chars and sulfone groups (for the CH-TU); the weight loss of this step was around 17.93 and 13.5%, respectively. The total loss of both sorbents was in the range of 93.37 and 76.28%, respectively, which confirms the increase in hydrocarbons in CH-TU and reflects the successful grafting of thiourea. Siddiqi et al. [89] concluded that nitrogen-based materials are eliminated first at low temperatures, followed by another transition linked to organic skeleton degradation. The final weight loss for both sorbents was significantly different, at 93.37 and 76.28% for CH-F and CH-TU, respectively. This difference was due to the difference in ratios of hydrocarbons in both sorbents. The final residue of CH-TU was higher than that of CH-F because of the increase in the hydrocarbon skeleton via grafting of thiourea. Apparently, the grafting of thiourea significantly increases the thermal stability and shifts the temperature to higher values; this was apparent from the final loss temperature, which was around 560 and 700 °C for CH-F and CH-TU, respectively.

Figure S1 shows the comparison of DrTG of both sorbents. It is shown that the functionalization of thiourea changes the structure and shifts the temperature to higher values (306.22, 468.31, 544.21 and 673.5 °C) compared to the two peaks for CH-F (295.63 and 583.6).

3.1.3. Elemental Analysis and pHpzc

The elemental analysis shows an increase in the N content compared to the pristine sorbent, as well as the appearance of sulfur in the final product. The N content was almost exactly the same, and shows an increasing ratio from 2.93 to 4.45 mmol, while S elements were noted in the final product (functionalized sorbent), with a molar ratio of only 1.28 mmol. These results verify the successful grafting of thiourea (considered the main source of S elements and supporting the increase in the N content). The full data from the elemental analysis are reported in Table S4. Figure 3 shows the determination of pH_pzc using the pH-drift method; the determination was performed using 0.1 M NaCl as the background salt. The pH_pzc values were 6.38 and 5.64 for CH-F and CH-TU, respectively. It was concluded that in acidic and slightly acidic solutions, the sorbents’ surfaces were still positively charged (at a pH below 6.38 and 5.64 for CH-F and CH-TU, respectively). The functionalized and non-functionalized sorbents with formaldehyde exhibit clear changes in their acid–base properties, leading to a shift in global pH_pzc.

3.1.4. SEM Analysis

Figure 4a–d show the scanning electron microscopy images and the size of the particles. Both sorbents seem to have the same external shape, which is characterized by an elongated structure; the majority of the particles had an unsymmetrical structure. It was noted that the size decreased after functionalization from an average of 19 µm to 14 µm for CH-F and CH-TU, respectively.

3.1.5. BET Surface Area

The particle size of the sorbents was investigated using N₂ sorption/desorption isotherms. The SSA (specific surface area) was close to 45–48 m² g⁻¹ for CH-F, while it increased to 52–54 for CH-TU. The porous volume appeared to be higher for CH-TU than for CH-F (≈0.715 cm³ g⁻¹ vs. 0.52 cm³ g⁻¹, respectively), while the pore size decreased from 572–512 Å to 466–435 Å, respectively. The largest pore size highlighted the possible microporous nature of the sorbents.

3.2. Silver Sorption

3.2.1. Effect of pH

The binding mechanisms of metal ions with reactive functional groups grafted on the sorbent surface are controlled by the pH of the medium. The pH had a direct effect on the speciation of metal ions and the nature of the functional groups (protonation/deprotonation properties). Silver species show a poor effect toward the pH (Figure S2 shows the speciation diagram of Ag(I); precipitation was detected over pH 9.3). From Figure 5, the sorption began with a low capacity, followed by an increase until it stabilized at a pH around 6. At acidic pH values (1 and 2), electrostatic repulsion was present between the protonated functional groups and positively charged metal ions. With an increase in the pH (over 3), the positive charge on the functional groups decreased and consequently so did the repulsion effect, causing an increase in the loading capacity. The binding was favorable for cation exchange with amine groups (from chitosan and thiourea) and thiol groups (obtained from tautomerization of thion groups). The small deviation in the error bars emphasizes the good reproducibility of the sorption experiments.

Under the selected experimental conditions (T: 21 °C and pH_eq: 6.4), the sorption capacities of both sorbents reached 0.549 and 1.31 mmol g⁻¹ for CH-F and CH-TU, respectively. This means that the sorption increased 2.5-fold as a result of functionalization by thiourea. The preference of thion (thiol groups) for silver can be explained by the hard and soft acids and bases theory (in which the silver ions are considered as soft acids which are attracted to a soft base (thiol groups)) or the S-based ligands [42]. Figure S3 shows the variation in the pH after sorption, which indicates that the two sorbents had the same profile at acidic pH (slightly increased), and the opposite trend was shown in slightly acidic or alkaline media. It was shown that after pH 6, CH-F slightly increased, while CH-TU slightly decreased; ∆pH ≈ 0.4 pH.

3.2.2. Uptake Kinetics

Several profiles were used for the uptake kinetics, i.e., resistance and diffusion rate, which include the film, intraparticle and bulk diffusion and proper reaction rate [90] in the batch system. The bulk reaction rate and resistance to film diffusion were minimized by the agitation speed (210–220 rpm (the sorption condition)), which is sufficient to minimize these kinds of profiles. Therefore, the analysis of kinetic profiles was performed with the PFORE (pseudo-first-order rate equation), PSORE (pseudo-second-order rate equation) and RIDE (resistance to intraparticle diffusion), which is the so-called Crank equation. Table S1 reports the relevant equations and the parameters of each one. In Figure 6, the faster sorption kinetics of the functionalized sorbent are shown compared to the non-functionalized one. Total saturation of CH-TU and CH-F was achieved within 30 and 40 min, respectively. The PFORE fitted the experimental profiles for both sorbents, in which the initial slopes of both sorbents follow the same trend. The low values of the error bars confirm the high reproducibility of the experimental test.

The fast kinetics reveal the easy diffusion of Ag(I) ions inside the pores.

Table 1. Average values of the three repeated experiments for the uptake kinetics of Ag(I) sorption using CH-F and CH-TU sorbents.

Model	Parameter	Unit	CH-F	CH-TU
	qeq,exp	mmol Ag g−1	0.551	1.342
PFORE	qeq,1	mmol Ag g−1	0.549	1.358
	k1 × 10	min−1	0.559	0.731
	R2	-	0.997	0.995
	AIC	-	−105	−118
PSORE	qeq,2	mmol Ag g−1	0.789	1.819
	k2 × 10	gmmol−1 min−1	1.55	1.94
	R2	-	0.493	0.512
	AIC	-	−13	−15
RIDE	De × 108	m2 min−1	1.95	2.03
	R2		0.38	0.53
	AIC		−15	−19

reports the three models’ parameters. The PFORE was used through comparable matching of both the calculated and experimental values of the sorption capacities and other statistic parameters (Akaike information criterion (AIC) values and R²). The PSORE (dotted lines) and RIDE (dash lines) fit the experimental values at low time values, but failed for higher time values; the PFORE (solid lines) fits for all time values.

Several contributions show how it is risky to discuss and interpret PFOREs and PSOREs that are unfit for the experimental parameters and the model requisites [91,92].

3.2.3. Sorption Isotherms

Figure 7 shows a comparable study of the two sorbents using the sorption isotherms (q_eq = f(C_eq)). The sorption properties were improved by grafting thiourea moieties (increased by three-fold) under the selected experimental conditions (pH₀ 6 at room temperature (21 ± 1 °C)). The maximum sorption capacities reached 0.763 mmol Ag g⁻¹ and 2.125 mmol Ag g⁻¹ for CH-F and CH-TU, respectively. The expected binding mechanism of Ag(I) ions in CH-F occurred through the amine and hydroxyl groups (which are considered as hard base groups), while in CH-TU, the thiol shared the binding with the already existing functional groups (through tautomerization of thion groups). Sulfur-based materials are considered as a soft base and favorable for Ag(I) (a soft acid). The three models are shown in Figure 7. The Freundlich equation (dash lines; with a bad fit) is known as a power-type equation. The Langmuir equation (solid lines) describes monolayer sorption without sorbed interactions, with an expected homogeneous sorption energy, and the Sips equation (dotted line; combination of Freundlich and Langmuir equations; the third parameter is used for adjusting and improving the mathematical quality of fits) was considered a good fit for the experimental data.

Table 2. Sorption isotherms of Ag(I) using CH-F and CH-TU sorbents.

Model	Parameter	Unit	CH-F	CH-TU
	qm,exp	mmol Ag g−1	0.763	2.125
Langmuir	qm,L	mmol Ag g−1	0.779	2.135
	bL	L mmol−1	1.11	3.24
	R2	-	0.995	0.993
	AIC	-	−138	−134
Freundlich	kF	L1/nF mmol1−1/nF g−1	1.53	1.87
	nF	-	2.33	2.14
	R2	-	0.754	0.562
	AIC	-	−29	−23
Sips	qm,S	mmol Ag g−1	0.792	1.227
	bS	(L mmol−1)1/nS	2.11	2.35
	nS	-	0.982	1.15
	R2	-	0.992	0.991
	AIC	-	−113	−122

reports the parameters (average values of the three repeated experiments) of the used models. By comparing the statistical criteria, fitting quality and AIC values, we found that the sorption isotherms were better fitted by the Sips and Langmuir equations than the Freundlich equation.

Table 2 reports the results of the comparison studies of the parameters of the three equations; it was noted that for the Langmuir and Sips equations, the overestimation did not exceed more than 10% of the calculated value. The Langmuir affinity coefficient, b_L, shows high values after functionalization (CH-F (1.11 L mmol⁻¹) < CH-TU (3.24 mmol⁻¹)); these data were significantly similar to those of the Sips equation; i.e., n_S increased to a small extent from 0.982 to 1.15 for CH-F and CH-TU, respectively.

The expected sorption mechanisms of Ag(I) and the reactive functional groups were determined through the data collected from FTIR spectra (through the functional groups that are used in the sorption, emphasizing the formation of thiol groups through tautomerization of the thion groups in the composite); through pH_PZC, which shows the overall charge of the sorbent; and through the metal speciation of Ag(I), as in Figure S2 (which provides two species at the experimental pH: free cationic Ag⁺ species and an aqua complex (Ag(H₂O)₄⁺)). These data indicate variable values for both sorbents; the maximum sorption pH was below the pH_pzc point for CH-TU and over it for CH-F. Mixed mechanisms were predicted using the chelating properties (the deprotonated functional groups) of the completely negatively charged functionalized sorbent and the partially positively charged non-functionalized one with the positively charged metal ions (either as free cationic species or as a partially dehydrated aqua complex). At this pH (optimum sorption), electron pairs on the chelating groups will be available for binding with positively charged metal ions, indicating mainly chelation for CH-TU and a mix of chelation and ion exchange for CH-F, as shown in Scheme 4.

Table 3. Comparison of the sorption properties of Ag(I) with some alternative sorbents (T = 20–25 °C).

Sorbent	pH	Time	qeq,exp.	qeq,L.	bL	Ref.
Biocomposite hydrogel (CMCellulose/CMChitosan/NaSulfon)	5	2880	0.0033	0.0042	205	[96]
Alginate-functionalized amino-carbamate beads	5	180	1.28	1.95	0.503	[97]
Poly-pyrrole thiol functionalization	5.63	480	7.23	7.48	150	[95]
Cellulose/L-cysteine microsphere	n.d.	540	0.584	0.618	17.4	[23]
Sargassum biomass (Ca-loaded)	5	1440	0.6	0.935	0.317	[28]
Sulfur-bearing resin	1	360	-	1.16	6.47	[98]
Dithiocarbamate-modified cellulose	0.2 M HNO3	70	-	9.94	24.9	[94]
Chitosan grafted with mercapto-benzimidazole	6.8	60	-	2.02	1.05	[35]
Nanoscale iron0/activated carbon (a)	5–6	40	9.15	12.0	0.649	[93]
Supercapacitor-activated carbon	4.5	60	2.16	2.45	0.583	[99]
Supercapacitor recycled AC	4.5	60	0.964	1.03	0.683	[99]
Gracilaria rhodophyta biochar	5.7	45	-	0.0013	244	[100]
Magnetic chitosan trione-pyrimidine derivative (MC-PYO)	6	90	1.91	2.15	2.33	[16]
Magnetic chitosan trithione-pyrimidine derivative (MC-PYS)	6	60	2.33	2.38	6.68	[16]
Magnetic chitosan (MC)	6	90	0.678	0.844	0.870	[16]
Thiourea–PVDF composite	6	360	1.60	1.86	40.4	[41]
Straw biochar	5	60	0.25	0.249	108	[101]
Cysteine/cellulose/thiol-ene	3	120	-	0.222	490	[34]
Thiol/graphene oxide	5	240	1.25	1.24	189	[102]
Thiosemicarbazide-based cellulose	7	2160	0.417	0.476	18.7	[103]
Hydroxyapatite	5.9	60	-	0.173	n.d.	[104]
CH-F	6	35	0.763	0.779	1.11	This study
CH-TU	6	20	2.125	2.135	3.24

q_eq,exp. (mmol g⁻¹): experimental sorption capacity, q_eqL, (mmol g⁻¹): sorption capacity at equilibrium according to the Langmuir equation, b_L (L mmol⁻¹): affinity coefficient; ^(a): occurrence of reductive precipitation; n.d.: not documented; units: time, min; q_eq, mmol Ag g⁻¹; b_L, L mmol⁻¹.

reports comparison studies from the literature of CH-F and CH-TU sorbents. Some materials have high sorption capacities, such as dithiooxamide coupling with formaldehyde (30.9 mmol Ag g⁻¹) [22], nanoscale iron⁰/activated carbon (q_m,L: 12.0) [93], dithiocarbamate-modified cellulose (q_m,L: 9.94) [94], poly-pyrrole thiol functionalization (q_{m, L}: 7.48) [95], and sorbents derived from thiourea (i.e., a thiourea–PVDF composite shows a relative decrease in the sorption (q_m,L: 1.86 mmol g⁻¹)) [41].

3.2.4. Sorption Selectivity

Acidic effluents contain diverse competitive ions, which affect the sorption performances of the prepared sorbents. This test was conducted to estimate the impact of polymetallic components on the sorption efficiency of the target metal ions. Both sorbents were tested for Ag(I) sorption in an equimolar solution with 1 mmol L⁻¹ of Na(I), Mg(II), Ca(II), Fe(III), Al(III) and Cu(II). The sorption experiments were performed at different pH values (ranging from 3 to 7). Figure 8 shows the effect of pH on the selectivity ratio of CH-F and CH-TU for Ag(I) against other metal ions. SC_Ag/Metal was calculated using the below equation, Equation (3):

$${SC}_{M1/M2}=\frac{D_{M1}}{D_{M2}}=\frac{q_{eq,M1}\times C_{eq,M2}}{C_{eq,M1}\times q_{eq,M2}}$$

(3)

From this figure, it was concluded that grafting of thiourea not only increases the loading performance but also the selectivity for multiple metal ions. From Scheme 4, both sorbents have comparable contents of NH and OH, while the functionalized sorbent contains thion groups (derived from the grafted thiourea that is produced by tautomerization), which convert to thiol groups that favor Ag(I) sorption, as discussed before.

The selectivity of both sorbents was higher at a pH_eq of 6.89 and 6.71 for CH-F and CH-TU, respectively, and had the following order:

• For CH-F/pH_eq 6.89: Ca(II) ≈ Na(I) > Mg(II) > Fe(II) > Al(III) ≈ Cu(II) > Ag(I).
• For CH-TU/pH_eq 6.71: Ag(I) > Ca(II) > Mg(II) >Fe(III) > Al(III) > Na(I) ≈Cu(II).

This is highlighted in Table S5 for both sorbents. The preference towards Ag(I) and Cu(II) is due to their classification as soft/intermediary metals. It was noteworthy that S-based ligands are considered as a soft base, which have higher ability to bind to soft ions (such as Ag⁺ or Cu²⁺), while O-based ligands are considered as a hard base and more attracted to a hard metal character for binding (i.e., Ca²⁺, Na⁺, and Mg²⁺) [42].

3.2.5. Metal Desorption and Recycling Performance

Reuse of the sorbents is considered an important property. The highest sorption was recorded at pHs higher than 5, which allows the lone pair of electrons to bind with cationic ions through free protons on the amines and hydroxyls. A 0.3 M HNO₃ solution was a good tool for the elution of all the adsorbed metal ions. Figure 9 shows the elution efficiency in nitric acid, which was estimated by the following equation (Equation (4)).

Elution efficiency = (C_(t) × V)/(q_max × Wt) × 100

(4)

where C_(t) is the concentration of metal ions in the solution at time (t), V is the volume of the eluent, q_max is the maximum loading capacity and Wt is the weight of the sorbent.

It is clear that the maximum desorption time was around 30 and 15 min for CH-F and CH-TU, respectively, which was sufficient for complete elution.

After desorption, the sorbents were rinsed with water for another loading for five cycles.

Table 4. The recycling data for CH-F and CH-TU sorbents.

	CH-F				CH-TU
Cycle	SE	StD	DE	StD	SE	StD	DE	StD
1	12.491	0.227	100.00	0.639	65.769	0.868	100.00	0.635
2	12.072	0.337	100.00	0.979	65.172	0.393	100.00	0.437
3	11.706	0.579	99.99	0.006	64.797	0.265	100.00	0.183
4	11.422	0.263	100.00	0.079	64.537	0.131	99.98	0.272
5	11.172	0.41	100.00	0.12	64.162	0.276	100.04	0.209

SE: sorption efficiency (%); DE: desorption efficiency (%); StD: standard deviation.

reports the efficiency (adsorption/desorption) over the five cycles. CH-F and CH-TU show a relative stability in the recycling processes, in which CH-TU is more efficient and has a high stability in the recycling process. The desorption of both sorbents reached 100%, while the losses in sorption performances were around 5 and 3% for CH-F and CH-TU, respectively.

3.3. Application to a Real Effluent—Acidic Leachate of Waste Photographic Film

Another test was performed on a polymetallic complex solution from acidic leaching (from 7 M nitric acid for 3 h at 50 °C) of reprocessed photographic films. The leaching solution consists of a high concentration of Ag(I) compared to other elements. The produced leaching solution had a wide content of the investigated elements—Fe, Cu, Mg, Cr, Co, Al, Ag, and Ni—with concentrations of 7.92, 1.53, 1.02, 0.71, 3.96, 2.3, 15.02, and 2.51 mg L⁻¹, respectively. Sorption was achieved via the agitation method with an SD of 0.66 g L⁻¹ at two different pH values (2 and 6) for both sorbents for around 10 h with an agitation speed of 210 rpm. After sorption, the sorbents were washed with water for neutralization and the metal ions were completely removed from pores; then, the elution step was performed. Desorption of the adsorbed metal ions was achieved using 0.3 M HNO₃. The outlet solution after adsorption was measured and the sorption selectivity is reported in Figure 10. A high adsorption activity and selectivity were noted at a high pH compared to those at low pH values. The selectivity is shown in Table S6; the order is as follows:

• CH-F/pH_eq/2.35: Al > Cr > N i> Co> Fe > Mg ≈ Cu; CH-F/pH_eq/6.11: Ni > Al > Co ≈ Cr ≈ Mg > Fe ≈ Cu. For CH-TU/pH_eq/2.19: Al > Co ≈ Cr > Fe > Mg > Ni> Cu; CH-F/pH_eq/5.86: Ni > Mg > Co> Al > Cu> Cr > Fe.

4. Conclusions

Chitosan particles were successfully functionalized using thiourea moieties with formaldehyde (which acts as a coupling agent). The produced functionalized sorbent was characterized by a high affinity for silver sorption compared to the non-functionalized sorbent. The sorbents were characterized via FTIR, SEM-EDX, TGA, EA and BET surface area. The adsorption kinetics were fitted using the PFORE (with equilibrium at 40 and 30 min for CH-F and CH-TU, respectively), while the isotherms were identical to the Sips and Langmuir equations. The maximum adsorption for Ag(I) was achieved at pH₀ 6, with a capacity of 0.763 and 2.125 mmol Ag g⁻¹ for CH-F and CH-TU, respectively. The outcome of the sorption performance indicates the significant effect of thiol grafting. Tests of polymetallic solutions (either equimolar synthetic or waste leaching solutions) show the preference of CH-TU for Ag compared to CH-F against other competitive metal ions. This is because S-based ligands have a higher affinity for Ag (i.e., a soft acid) than O- and N-based ligands (which exhibit the opposite effect (a preference for hard acids)). The sorbent shows a high affinity for adsorption of Ag(I) compared with other values found in the literature. Nitric acid solution is considered a good eluent for the desorbed metal ions, and both sorbents show a high stability against sorption–desorption cycles, with the limit decreasing with the affinity (around 5 and 3% for CH-F and CH-TU, respectively). Finally, the functionalized sorbent shows a high affinity for recovering silver ions from acidic waste solutions. Moreover, the modified biosorbent was synthesized in a one-pot process via a green synthesis method (using water instead of solvent), and it shows a significant selectivity for Ag(I) over several ions (cations and anions) in a poly-metallic acidic leaching solution. This study provides a novel and eco-friendly sorbent for efficient Ag⁺ recovery, reflecting its potential application in environmental monitoring, where it can be reused for long-term practical applications.